DeepSeek V4 Pro
Recommended
Model source: Hugging FaceThe most capable in the catalogue: reasons through long problems, writes code and drives your tools.
Reads in one go
≈ 2,200 pages
1,048,576 tokens
Brain size
The largest in the catalogue
1.6T MoE, FP8
Where it runs
Spain (EU)
8× B200
Understands
Text, not images
Chat, Coding, Reasoning, Tools
What it is good at
- Reasoning. Thinks in several steps before answering, for problems that take more than one.
- Coding. Understands large codebases and proposes complete changes.
- Tools. Searches, queries data and calls other systems. The basis of an agent.
- MoE. Mixture of Experts: it splits big tasks and shares them among specialists, switching on only the ones it needs. It performs like a huge model without costing like one.
- Long context. Reads enormous documents in one go without losing the thread.
Only need simple chat, classification or translation? Qwen3.5 9B does it for a fraction of the price, and switching model is switching one word.
Pricing
€ per million tokens, VAT included
You send (input)0.80 €
It answers (output)2.60 €
Re-reading repeats (cache)0.08 €
1,000 chat answers~1.18 €
Analysing a 200 page contract~0.08 €
try it with 50 € freeNo card. Enough for about 42,373 answers.
For your technical team
Compatible with the OpenAI SDK: change the base URL and the key, and that is it.
curl https://api.gpuflow.ai/v1/chat/completions \
-H "Authorization: Bearer $GPUFLOW_API_KEY" \
-d '{ "model": "deepseek-v4-pro",
"messages": [{ "role": "user", "content": "Hola" }] }'